Looking at regression results for how much a respondent supports elections over services based on the number of household items they have and their education level.
| support elections over services | ||
| (1) | (2) | |
| household items | 0.019*** | 0.024*** |
| (0.002) | (0.002) | |
| education | 0.034*** | |
| (0.007) | ||
| Constant | 1.297*** | 1.348*** |
| (0.014) | (0.011) | |
| Observations | 12,556 | 12,585 |
| Adjusted R2 | 0.014 | 0.012 |
| Note: | p<0.05; p<0.01; p<0.001 | |
Simple linear regression: using only household items to predict support for elections instead of services. \[\begin{align*} \text{elections over services} & = \hat{\beta}_1 * \text{household items} + \hat{\beta}_0 + \epsilon \\ \text{elections over services} & = 0.024 * \text{household items} + 1.348 + \epsilon \end{align*}\]
Multiple linear regression: controlling for education levels.
\[\begin{align*} \text{elections over services} & = \hat{\beta}_1 * \text{household items} +\hat{\beta}_2 * \text{education} + \hat{\beta}_0 + \epsilon \\ \text{elections over services} & = 0.019 * \text{household items} +0.034 * \text{education} +1.297 + \epsilon \end{align*}\]
Adding colors for countries. This is messy, but fun to explore. Hover over a dot to see which country that respondent lives in.